In the past several years, a new and vigorous branch of virology ("molecular pathogenicity") has developed which deals with the molecular events that occur when viruses interact with the antiviral defenses of the immune system. A significant development in this field was our discovery that two "sets" of proteins, the 14,700 MW (14.7K) protein and the 10.4K/14.5K complex of proteins, coded by the E3 transcription unit of adenovirus, function independently to prevent cytolysis by tumor necrosis factor (TNF). 14.7K, and possibly 10.4K/14.5K, also affect aspects of TNF signal transduction. TNF, which is secreted by activated macrophages, is a key mediator of the inflammatory and immune responses, and it inhibits virus replication in cultured cells. Thus, TNF is an important antiviral defense of the host. We hypothesize that these adenovirus E3 proteins protect the infected cell in the host from cytolysis by TNF, and probably also from the inflammatory response amplified by TNF. We propose to continue our studies on how these proteins prevent TNF cytolysis, and how they affect TNF signal transduction in general. Little is known about these topics, and the E3 proteins should be unique and powerful sources of information. TNF kills most cells by apoptosis, so the E3 proteins should provide insights into this important and poorly understood topic as well. Excessive TNF in humans is associated with cachexia, AIDS, and sepsis, and TNF is being used in trials to treat cancer. Information gleaned from our studies may be useful in understanding and treating not only virus infections but also cancer. Studies on these E3 proteins will be conducted using viruses and also with the E3 genes expressed from inducible vectors. 14.7K, purified to near homogeneity, will be injected into cells to address whether it can prevent TNF lysis with or without protein synthesis. 14.7K has been shown to prevent activation of phospholipase A2 by TNF, and to affect the ability of TNF to induce manganese superoxide dismutase; studies are proposed to address the mechanism by which this occurs, and if 10.4K/14.5K have similar functions. We will also address whether 14.7K and 10.4K/14.5K prevent apoptosis induced by the anti-Fas or Apo-1 monoclonal antibodies, by deprivation of growth factors, or by infection with HIV, and whether the E3 proteins prevent apoptosis via the Bcl-2 or p53 proteins. Several cellular proteins that bind to 14.7K will be characterized in detail. Virus mutants with in-frame deletions and point mutations in the 14.7K and 10.4K genes will be studied to develop a structure-function map of these proteins.